Test types

What it is

AreaCacao supports 11 sensory test types, the most widely used in research, quality control, and panel training, based on international standards. You pick the type in Step 1 of the wizard when creating a tasting (see Create a tasting).

The wizard groups the tests into 4 blocks. When you click a type, the wizard advances to the next step on its own.

How to read this page

For each test you'll find:

• What it measures and when to use it.
• The reference standard, when one applies.
• How the leader configures it from the wizard.
• How the assessor answers it on their phone (see the Assessor guide for the screen-by-screen detail).

Block 1 — Most common

This is the block opened by default in the wizard.

Descriptive tasting

• What it measures: the panel describes the sensory profile of each sample on a 0–10 scale per attribute. The results view gives you the spider chart and the ANOVA/Tukey table; the color wheel is generated in the technical dossier (PDF), not on the results screen.
• When: to characterize a product, compare variants, or run batch-to-batch quality control with a trained panel.
• How the leader configures it: pick Descriptive tasting, define the samples, and select an attribute protocol in Step 3.
• How the assessor answers it: they evaluate sample by sample. For each one they score the intensity per attribute (0 = absent, 10 = maximum), mark Defects detected, give an Overall rating of the sample, and add Free notes.

The module sidebar has a Descriptive tasting shortcut that opens the wizard straight at the Samples step with this type preselected.

Triangle test

• What it measures: whether there is a perceptible sensory difference between two products. 3 coded samples are presented (2 identical and 1 different) and the panel identifies the different one. The system computes the binomial p-value (chance hit = 1/3).
• Standard: ISO 4120.
• When: to validate a supplier change, a recipe modification, or a process change.
• How the leader configures it: pick Triangle test and define the two versions to compare.
• How the assessor answers it: a single holistic screen. They taste the 3 samples in order and mark which one is "the different one"; submission is one-shot.

Hedonic (consumer)

• What it measures: the preference or acceptance of real consumers. It uses the 9-point Peryam scale with full verbal labels, from "dislike extremely" to "like extremely".
• Standard: ISO 11136.
• When: studies with end consumers, not an expert panel. It's the go-to test to send as a public link to anonymous consumers, since it doesn't require a trained panel. The public link and QR code also work for other test types.
• How the leader configures it: pick Hedonic (consumer) and assign panelists of the consumer or guest type. The hedonic test uses the 9-point Peryam scale directly: it no longer asks you to pick an 18-attribute descriptive protocol, so the wizard skips the Protocol step (the flow has 6 steps instead of 7).
• How the assessor answers it: sample by sample, picking their position on the vertical 1–9 scale (Overall hedonic rating).

Block 2 — Other comparative tests

Paired comparison (2-AFC)

• What it measures: directed preference or intensity between 2 samples, with a specific question (for example, "which is sweeter?"). Chance hit = 1/2.
• Standard: ISO 5495.
• When: a directed comparison with a clear criterion.
• How the leader configures it: pick Paired comparison (2-AFC) and write the Directed question in the logistics.
• How the assessor answers it: a single holistic screen; they pick one of the two samples and confirm.

Duo-trio

• What it measures: discrimination that's gentler than the triangle test. A reference + 2 coded samples are presented; the panel identifies which of the two is different from the reference.
• Standard: ISO 10399.
• When: fillings or subtle-profile products where the triangle test is too demanding.
• How the leader configures it: pick Duo-trio and define the reference and the two samples.
• How the assessor answers it: a single holistic screen; they taste the reference and the two coded samples, and mark the different one.

Simple ranking

• What it measures: the order of several samples by an attribute (intensity, sweetness, acidity). The system analyzes it with the Friedman test and you get the ranking report. Ranking requires 3 to 8 samples (ISO 8587).
• When: to establish a ranking by a specific property.
• How the leader configures it: pick Simple ranking and write the Directed question that defines the ordering criterion.
• How the assessor answers it: a holistic screen; they order all the samples and submit.

Ranking with scales

• What it measures: the same as simple ranking, but the panelist also assigns a 0–10 intensity to each sample. This enables statistical analysis on the scores.
• When: when you need the ranking and, at the same time, the distance between samples.
• How the leader configures it: pick Ranking with scales.
• How the assessor answers it: a holistic screen; they order the samples and score the intensity of each one.

Block 3 — Advanced tests

This block is collapsed by default and shows the notice "For a trained panel. They demand time and rigorous method."

Time-intensity

• What it measures: the perception dynamics of an attribute over time. After introducing the sample, the assessor drags a continuous slider for 60 seconds; the system reconstructs the intensity-vs-time curve.
• When: to study how an attribute evolves (sweetness that drops, lingering bitterness, aftertaste) and not just its average intensity.
• How the leader configures it: pick Time-intensity and define the attribute to track.
• How the assessor answers it: sample by sample. They introduce the sample and move the slider (from "absent" to maximum) as the 60 seconds elapse; the curve is recorded automatically.

Aroma release

• What it measures: how the sample's aroma is released at fixed intervals (0s start, 10s attack, 20s development, 40s plateau, 60s decline, and 120s after-aroma). It's the sensory version of a release study, without lab equipment.
• When: to characterize the aromatic evolution of a product over the course of consumption.
• How the leader configures it: pick Aroma release and define the aromatic attributes to track.
• How the assessor answers it: sample by sample. The device alerts them when each interval is reached and the panelist scores the 0–10 intensity of the indicated aromatic attributes.

Block 4 — Panel training

This block is collapsed by default and shows the notice "Qualify your panelists per ISO 8586. Results feed the panelist profile (naive → initiated → selected → expert)."

In these two tests the wizard skips the Protocol step: the method is already fixed by the standard.

Basic taste recognition

• What it measures: that the panelist correctly identifies the basic tastes. The test serves 5 or 6 solutions (sweet, salty, sour, bitter, umami, and, optionally, metallic) in random order, and returns a score of X/6.
• Standard: ISO 3972:2011.
• When: the initial calibration of a panelist and periodic revalidations.
• How the assessor answers it: for each solution they identify the perceived taste.
• See Basic taste recognition.

Threshold detection

• What it measures: the panelist's detection threshold for a stimulus, via a 3-AFC series with increasing concentrations. The system computes the Best Estimate Threshold (BET).
• Standards: ISO 13301:2018 + ASTM E679-19.
• When: to move a panelist from selected to expert with quantitative backing.
• How the assessor answers it: at each level they taste 3 presentations and indicate which is the different one; the hit and the concentration are resolved on the server.
• See Threshold detection.

Summary table

Every classic tasting requires a minimum of 2 samples (4–6 recommended) and accepts up to 20. The exception is ranking, which requires 3 to 8 samples (ISO 8587). Training tests aren't configured by samples, but by solutions or concentration levels.

Related terms